OpenFAST
Bug report: Lack of Data Sanity Checks and Write Outputs for TLCDs in Structural Control (StC)
Bug description
I looked briefly at the source code and don't see that there are any data sanity checks for tuned liquid column dampers (TLCDs) in SUBROUTINE StC_ValidatePrimaryData() of source file StrucCtrl.f90. Without such sanity checks, you can specify inputs that lead to an ill-conditioning of the equations, e.g., division by zero if a TLCD mass is zero. This issue should be fixed at some point.
Also, there are no write outputs specific to TLCDs, such as liquid displacement, w, that would be useful to track the status of the TLCD. Such outputs should be added.
To Reproduce
Enable an StC in ServoDyn, and in the StC input file, set StC_DOF_MODE = 3, but set the properties of an unwanted x- or y-based TLCD to zero. Because the mass of the unwanted TLCD is zero, a division by zero error will trigger NaNs, leading to a simulation abort.
Expected behavior StC should warn if the TLCD inputs are ill defined. It may also be useful for the the user to specify whether the x- or y-based TLCD are enabled/disabled independently, like they are for tuned mass dampers (TMDs).
Write outputs should also be added for TLCDs.
OpenFAST Version Any version since the StC submodel of ServoDyn was first introduced.
Additional context This issue was first raised on our forum: https://forums.nrel.gov/t/problems-when-using-tlcd-in-nrel5mw-oc4/9010.
Until this is fixed, users should ensure that the mass of an unwanted TLCD is near zero, but not exactly zero, by ensuring some nonzero values of TLCD length, area, and density.
Additional context Furthermore, I also hope that someone can also review the formula applied in the source code for assessing the mass of the TLCD application in the structure, and include Mass of TLCD in the outputs...
The forum link also includes the issue of maintaining PtfmHeave by removing equal amounts of ballast water.
When Z-axis symmetric TMDs are attached to the NREL 5MW - OC4 semi-platform or just placed on the Z-axis, using a similar formula to maintain the waterline is useful. However, for TLCD, it was observed that the larger the mass, the greater the difference.
Details are also showed on the forum: https://forums.nrel.gov/t/problems-when-using-tlcd-in-nrel5mw-oc4/9010
